This was a randomized controlled feasibility study approved by an institutional review board (STUDY00011219; NCT04272138). Middle-aged adults with elevated stress levels (Textbox 1) were recruited for an “app-based health and well-being study,” via flyers encouraging men to participate shared on social media platforms (eg, Instagram and Facebook) and the ResearchMatch website [28]. All participants provided electronic consent. Eligibility, consent, demographic, and survey data were collected using REDCap (Research Electronic Data Capture), a secure, web-based software platform, hosted by Arizona State University [29,30].

Participants were randomized to an app-based meditation intervention (Calm) or an app-based education control group (POD, an app that delivered podcasts on health and well-being, but excluding mindfulness, stress, or sleep content in the same context that a consumer-based mindfulness meditation app delivers content [31]) by using a randomized numbered list generated through simple randomization via the Research Randomizer website [32]. Participants were asked to complete 10-minute meditations daily on the Calm app or to listen to 10-minute educational podcasts daily on the POD app, for 4 weeks. All participants received weekly reminders and links to REDCap assessments via email.

Participants completed self-report assessments at baseline and postintervention (ie, week 4). Feasibility was measured as acceptability and demand using the Bowen framework (Table 1) [27], and feasibility benchmarks were established using a previously established methodology of feasibility trials with Calm [25,33]. Benchmarks for acceptability were more than 75% of participants reporting each of the following: satisfaction with the intervention, perceiving the app as appropriate and useful, and intent to continue using the app. Benchmarks for demand were more than 40% of the sample comprising male participants, more than 70% of the Calm app user group adhering to at least 70% of the meditation intervention (ie, ≥10 minutes/day of using Calm), and more than 75% retention of men allocated to the intervention group. Adherence was measured using objective app usage data collected by both Calm and POD. The measures used to explore preliminary effects and COVID-19 perceptions are listed in Table 2. The COVID-19 Perceptions Survey was added to the postsurveys in March 2020 following the start of recruitment.

Feasibility and COVID-19 perceptions data were examined using descriptive statistics. Preliminary effects on perceived stress, anxiety, depressive symptoms, mindfulness, coping, physical activity, and eating habits were examined using repeated measures analysis of variance. Data were analyzed in SPSS (version 26; IBM Corporation) and SAS (version 9.4; SAS Institute). Results were considered significant at a 2-tailed α value <.05.

In total, 83 middle-aged adults were consented and randomized, of which 60 (72%) were included in the analysis (Figure 1). Demographic characteristics of the sample are presented in Table 3.

No significant within or between group differences in stress or psychological outcomes related to stress were observed, nor were significant differences observed in health behaviors related to stress (Table 6).

Most participants in Calm reported that the COVID-19 pandemic affected their stress levels (26/28, 93%), mental health (23/28, 82%), and physical health (17/28, 61%) (Table 7).

A 4-week, app-based mindfulness meditation intervention (ie, Calm app) may be feasible for use among middle-aged adults. Calm group participants expressed satisfaction with the intervention and felt it was appropriate and useful. However, significant improvements in perceived stress and psychological outcomes (ie, anxiety, depressive symptoms, mindfulness, and general coping) or health behaviors related to stress (ie, physical activity and eating habits) were not observed among these participants. The majority of participants reported that COVID-19 has negatively affected their stress, mental health, and physical health.

We exceeded our benchmark (>75%) for acceptability rating of the Calm app among middle-aged adults experiencing stress. This finding was similar to that of other studies assessing the feasibility of Calm in patients with cancer and among college students [25]. Our benchmark for adherence to the meditation prescription (ie, >70% of the sample who completed at least 70% of the meditations) was met and better than most 4-week randomized controlled trials using an app to reduce stress [13,23,40]. Men had a higher adherence to the intervention than women (86% vs 71%), but this finding is not entirely consistent with other studies and does not necessarily suggest overall gender differences in meditation app use [41,42]. Research on app-based interventions targeting stress reduction and related outcomes, including objective app-usage data in middle-aged men, is warranted.

Our benchmark was to recruit 40% men, and we recruited 35% men (29/83) in our study. We were able to recruit more men than other app-based meditation studies (5.7%-27%), including studies that focused on middle-aged adults [21,25,43-46]. The retention of men (14/15, 93%) was also higher than that of women (19/24, 60%). Although, on average, men spent less time meditating than women did (approximately 57-91 vs 107-135 min/week) and completed fewer weekly sessions (4.7-7.5 vs 7.4-9.7 sessions/week), they were more likely to adhere to the 10-minute prescription. This finding suggests that Calm may be a useful self-management tool for both men and women to manage stress [21,44,45]. Future app-based meditation interventions should focus on recruitment and retention of men, especially because men are less likely to seek stress management strategies than women [47,48].

Significant changes in stress and related psychological outcomes or health behaviors were not observed. In another study testing the efficacy of meditation delivered via Calm, changes in stress and mindfulness in college students were observed after 8 weeks of participation. When taken together with these findings, the present data suggest that any significant changes in stress levels may take longer than 4 weeks to be noticeable [25].

Few studies have tested the effects of app-based mindfulness meditation on health behaviors related to stress [49] and have reported inconsistent findings [23,40,50,51]. We observed a negative trend regarding physical activity and eating habits. These may, however, have been related to lifestyle modifications due to the COVID-19 pandemic (eg, quarantine and closure).

Perceptions of COVID-19 could have had an important impact on stress in relation to the findings of this study. It is possible that meditation helped the study participants in the intervention group to maintain their stress and anxiety levels, as well as depressive symptoms (as opposed to elevated levels), during the COVID-19 pandemic [52]. Likewise, the general health education content of the control app may have helped mitigate the impact of the pandemic through avoidance or redirecting negative thoughts [53]. However, data regarding how participants felt their app usage impacted their COVID-19–related stress was not measured. More research on the effects of COVID-19 on stress and related health outcomes and how a meditation app may buffer that impact is warranted.

Limitations to this study include the following: (1) the majority of the sample comprised White participants, and generalizability of the findings may thus be limited; (2) the intervention was only 4 weeks long and did not include a follow-up; and (3) the COVID-19 pandemic may have had a notable impact on the findings of this study.

This study supports the feasibility of a 4-week, mobile app–based mindfulness meditation intervention (ie, Calm) in middle-aged men and women with specific application for the recruitment of men to inform future studies.